Phase I demonstrated the feasibility of an intra-oral artificial larynx controlled from the hand. Feasibility was proven by building and fitting prototype models to laryngectomees who have used the unit demonstrating improved naturalness and intelligibility. Although an improvement for laryngectomees has been achieved, additional experiments have shown that one cause of reduced naturalness in the unit is the location of the sound source in the mouth as opposed to the throat. A commercial device, Ultravoice(tm), resulted from Phase I. The next steps projected for Phase II work are: 1. an experiment with 40 laryngectomees to compare the Ultravoice (tm) unit with other speech techniques for laryngectomees; and 2. extended design work on an adaptive feedback system which would correct for the distortion caused by locating the sound source on the palate thereby improving sound quality and product value. PROPOSED COMMERCIAL APPLICATION: The Phase II work is expected to provide significant improvements in product quality, rehabilitation value and in the range of use.